mindroom

MindRoom documentation corpus for accurate product, configuration, and workflow guidance.

Run live end-to-end checks for this repository. Use when booting the local MindRoom stack or SaaS sandbox, starting an isolated local Matrix/backend instance, creating disposable Matrix accounts, sending or reading messages with Matty, hitting live API endpoints, taking frontend screenshots, or verifying behavior through the real UI instead of tests alone. Also use when a live-test workflow struggled and the skill itself should be improved.

Use when the user asks for native Codex sub-agents, parallel PR review agents, unbiased re-review loops, or main-thread fixes after agent findings in the MindRoom repository.

Use when fetching, triaging, or addressing PR review comments from GitHub, copied text, cloud AI reviewers, or other coding agents in the MindRoom repository.

Two coding agents debate code changes through a shared DEBATE.md file. Use when you want adversarial review of a commit, diff, or PR.

Use when spawning and supervising multiple agent-cli Codex worktrees through tmux, especially when you must poll progress, force prompt rereads against the local TASK file, run current-context and fresh-context pr-review loops, and verify that review fixes are actually pushed before merge.

Zero-tolerance pull request review. Every issue is a blocker. Use when reviewing PRs for merge readiness.

Review documentation for accuracy, completeness, and consistency against the actual codebase.

Find duplicated functionality across the codebase and propose minimal, safe generalizations.

Spawns AI coding agents in isolated git worktrees. Use when the user asks to spawn or launch an agent, delegate a task to a separate agent, work in a separate worktree, or parallelize development across features.